Vocal Tract Segmentation on 75-Speaker Annot-16 (SS-UT)
90.82DSC (%)Ours
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| OursMethod Type=Multimodal2026.05 | 90.82 | 1.48 | |
| OursMethod Type=Multimodal2026.05 | 86.37 | 2.36 | |
| Jain et al.Method Type=Multimodal2026.05 | 85.35 | 2.21 | |
| AVSegFormerMethod Type=Multimodal, Adapted=true2026.05 | 83.68 | 2.8 | |
| AV-SAMMethod Type=Multimodal, Adapted=true2026.05 | 82.27 | 2.38 | |
| U-MambaMethod Type=Unimodal2026.05 | 81.71 | 2.52 | |
| AV-SAMMethod Type=Multimodal, Adapted=true2026.05 | 81.21 | 3.69 | |
| Jain et al.Method Type=Multimodal2026.05 | 81.19 | 5.54 | |
| AVSegFormerMethod Type=Multimodal, Adapted=true2026.05 | 80.53 | 3.48 | |
| U-MambaMethod Type=Unimodal2026.05 | 79.49 | 3.82 | |
| DINOv2Method Type=Unimodal2026.05 | 77.83 | 3.76 | |
| MedSAM-2Method Type=Unimodal2026.05 | 75.14 | 3.11 | |
| nnU-NetMethod Type=Unimodal2026.05 | 74.59 | 5.66 | |
| DINOv2Method Type=Unimodal2026.05 | 73.58 | 3.5 | |
| Swin-UNETRMethod Type=Unimodal2026.05 | 72.38 | 5.22 | |
| MedSAM-2Method Type=Unimodal2026.05 | 71.57 | 3.5 | |
| Swin-UNETRMethod Type=Unimodal2026.05 | 70.61 | 4.37 | |
| nnU-NetMethod Type=Unimodal2026.05 | 69.77 | 6.87 | |
| ResUNet-aMethod Type=Unimodal2026.05 | 69.24 | 6.13 | |
| MedSegDiff-V2Method Type=Unimodal2026.05 | 69.01 | 5.1 | |
| ResUNet-aMethod Type=Unimodal2026.05 | 67.58 | 7.41 | |
| MedSegDiff-V2Method Type=Unimodal2026.05 | 65.43 | 6.71 |